• Air Toxics
• CMAQ
• Conformity
• It All Adds Up

• Laws
• Reference Material
• Research
• Publications
• Training
• Models & Methodologies

Freight Movement & Air Quality

Chapter 3: Freight Transportation Emissions at the Regional Level

The previous section discusses freight transportation and emissions at the national level. Yet air pollution is primarily a regional problem, affecting major metropolitan areas in particular. Moreover, freight transportation patterns are fundamentally different when viewed at the regional level compared to the national level. To examine the linkages between freight transportation and air quality at the regional level, we selected six major metropolitan areas for a focused look at freight emissions. These six metropolitan areas (Los Angeles, Dallas-Fort Worth, Houston, Chicago, Detroit, and Baltimore) were selected because they are all major multi-modal freight activity centers but reflect diversity in terms of geographic location, economic base, and freight modal balance.

This section presents estimates of freight transportation emissions in the six study areas. We first summarize freight transportation activity in the six regions, highlighting notable differences. Emissions are then presented by mode - trucking, freight rail, marine freight, and air freight. For each mode, we briefly discuss the methodology used to develop the emissions estimates, followed by the results. The last part of the section is a summary that compares emissions across all freight modes and regions. We generally defined the regions as the 1-hour ozone nonattainment or maintenance area; in the case of the Los Angeles region, we defined the region as the South Coast Air Quality Management District and Ventura County nonattainment areas.

In a typical regional emission inventory developed for a State Implementation Plan (SIP), emissions are presented by mode (e.g., on-road vehicles, rail, commercial marine, air), and no attempt is made to distinguish between freight and non-freight activity. The estimates in this section reflect emissions caused only by freight activity, as distinct from emissions from sources such as passenger rail, passenger ferries and other non-freight vessels, and passenger air transport. In the case of trucking, we have assumed that all heavy-duty trucks are involved in freight transportation, although we recognize that a small portion of these vehicles are engaged in service and construction activities that do not, strictly speaking, involve freight movement.

Note that in some cases, there is ambiguity as to what activity should actually be attributed to freight. For example, there are marine vessels like tugboats and airport ground support equipment that support both freight and non-freight movement. In these cases, we have attempted to apportion the activity and emissions caused by the freight sector as best as possible, while recognizing that available data do not allow precise apportionment between freight and non-freight. There is also equipment that supports the freight sector but does not directly involve freight movement, such as diesel-powered sweepers to clean port areas or equipment used to maintain railroad tracks. We generally did not estimate emissions from these sources because they contribute only negligibly to the freight sector total and because of limited data on the activity of this equipment.

3.1 Regional Freight Activity

All six regions profiled in this section are major U.S. freight centers, although there are some key differences in freight activity among the regions. Like every major U.S. city, trucking is the dominant freight mode in the six regions. Table 3-1 shows heavy-duty truck VMT in the six regions. The Los Angeles region has the most truck activity among the six - more than 7.8 billion VMT - followed by Chicago and Detroit.

As a percentage of total on-road VMT, Detroit ranks first among the six regions with 12.7 percent of VMT attributable to heavy-duty trucks. Chicago also has a high percentage of truck VMT (11.1 percent). These high fractions are probably due in part to a relatively large volumes of long-distance trucks passing into, out of, and through these regions, including U.S.-Canada truck traffic in the case of Detroit. The Los Angeles region has the smallest truck VMT fraction, most likely because the very large regional population and passenger vehicle activity means that trucks are responsible for a relatively small share of the on-road total.

Table 3-1: Heavy-Duty Truck VMT in the Six Study Regions, 2002

Source: Data provided by MPOs and state and regional air quality agencies.

While most freight trucking activity at the metropolitan level consists of movements within the region, freight movements by rail, marine, and air modes are dominated by inter-regional traffic. Table 3-2 shows freight tonnage into and out of the six study regions. These data show major differences among the six regions in terms of freight activity by mode. For example, rail freight plays a much more significant role in Chicago (36 percent of total intercity freight tonnage) than in the other regions (7 to 17 percent of tonnage). Chicago is the only city where all six major U.S. and Canadian Class I railroads come together to interchange freight. The region boasts 74 rail marshalling yards, including 17 intermodal terminals, substantially more than any of the other five study regions.

Table 3-2: Commodity Flows Into and Out of the Six Study Regions, 2003 (thousands of tons)

Source: FHWA Freight Analysis Framework (trucking and rail); Bureau of Transportation Statistics, Air Carrier Statistics T-100 database (air); U.S. Army Corps of Engineers, Waterborne Commerce of the United States database (marine).

Five of the six study regions have seaports (the exception being Dallas-Ft Worth). Waterborne freight activity is much greater in the regions with deep-sea ports (Los Angeles, Houston, Baltimore) than in the regions with Great Lakes ports (Chicago and Detroit). The Los Angeles and Houston area ports are among the largest in the nation, although they are quite different. The ports of Los Angeles and Long Beach are the leading container ports in the county, while Houston specializes in bulk petrochemicals and ranks near the top nationally on a tonnage basis.

Aircraft tend to be cost-effective only for small, high value shipments, so commodity flows by air make up only a fraction (less than 0.5 percent) of the total freight flows in every study region. In every region, the dominant passenger airport is also the dominant air cargo facility. Los Angeles International and O'Hare airports have by far the greatest air cargo activity among the six study areas. Two regions have secondary airports that are major air cargo facilities - Ontario Airport in the Los Angeles region and Alliance Airport in the Dallas-Ft Worth region.

3.2 Trucking Emissions

Trucking emissions are typically calculated as part of the total on-road vehicle emissions estimation process. Because on-road vehicles are one of the largest sources of pollutant emissions, and because of transportation conformity requirements under the Clean Air Act, the process used for estimating on-road vehicle activity and emissions is often more comprehensive and complex than for other transportation sources. All large metropolitan areas develop detailed estimates of VMT and on-road emissions by vehicle class and roadway functional class. For emission inventory purposes, most regions rely on the MPO travel demand forecasting model to determine VMT and vehicle speeds, calibrating the model to observed traffic counts. Other regions estimate VMT based directly on traffic counts.

Emission factors are developed using EPA's MOBILE6 model or, in California, the California Air Resources Board's EMFAC model. Development of the emission factors requires regionally specific information on inspection and maintenance (I/M) programs, fuel characteristics, temperature information, vehicle age distribution, and vehicle mileage accumulation by model year.

In most cases, we report the 2002 on-road inventory data developed by the MPO or state air quality agency for each region, typically developed as required by EPA's Consolidated Emissions Reporting Rule (CERR). In one case (Baltimore), a 2002 annual inventory was not available, so we estimated annual on-road emissions based on 2002 daily emissions calculated by the MPO for conformity purposes.³⁵

Summary of Methodology

All six study regions use a similar methodology to estimate on-road vehicle emissions. This methodology typically involves the following steps; some MPOs may perform these steps in a slightly different order.

1. The region's MPO uses a four-step travel demand model to estimate base year and future year traffic volumes by link. In some cases, the model estimates truck trips independent of passenger vehicle trips (i.e., independent truck trip generation and trip distribution modules). In other cases, the models estimate only passenger vehicle trips or total trips, and truck volumes are calculated as a percentage of the total volume (Step 5 below). Travel demand models use a computerized representation of the regional roadway system that includes all freeways and arterials but typically few or no local streets.
2. As required by EPA, the MPO adjusts the travel model traffic volumes based on observed traffic counts from the Highway Performance Monitoring System (HPMS), possibly supplemented with additional traffic counts. In this way, the model is calibrated to reflect base year conditions as accurately as possible.
3. The MPO estimates traffic volumes on local roads, which are not represented in a travel model. Some MPOs do this estimation themselves (e.g., the Baltimore MPO); others rely on local roadway VMT provided by the state DOT (e.g., the Detroit MPO).
4. Daily traffic volumes by link are disaggregated to hourly volumes, using observed traffic counts.
5. Model traffic volumes at the link level are allocated to major vehicle types, based on traffic count information. For example, BMC uses traffic count data provided by Maryland DOT to convert the hourly model traffic volumes into four vehicle types: motorcycles; 2-axle, 4-tire (passenger vehicles); buses; and 2-axle, 6-tire plus 3+ axles (heavy-duty trucks).
6. VMT is summed by vehicle type and facility type.
7. The MOBILE6 model requires VMT by 16 different vehicle types. The vehicle types are defined by vehicle configuration and the gross vehicle weight rating (GVWR). Most regions do not have VMT or traffic count information at this level of detail, so they rely on the MOBILE6 defaults to apportion VMT into these 16 vehicle types. In California, the EMFAC model uses only three weight classes for trucks (light-heavy duty, medium-heavy duty, and heavy-heavy-duty), each divided into gasoline catalyst, gasoline non-catalyst, and diesel engines.
8. Hourly speeds are estimated for each link. Because emission factors vary with vehicle speed, the distribution of VMT by speed can have an important effect on emissions. MPOs use equations that compare link-level volume and capacity to estimate speed. VMT is then grouped into 14 speed “bins.”
9. The distribution of VMT by speed will vary by roadway functional class. The four functional classes in MOBILE6 are freeway (excluding ramps), arterial/collector, local roadway, and freeway ramp. MOBILE6 allows users to enter a distribution of VMT by speed only for freeways and arterials/collectors. For local roadways and freeway ramps, the average speed in the model is fixed at 12.9 mph and 34.6 mph, respectively, and cannot be modified. Thus, MPOs will typically input a 24 x 14 x 2 matrix of VMT fractions (24 hours in the day, 14 speed bins, 2 facility types).
10. MOBILE6 input scripts are developed for information such as fuel Reid vapor pressure (RVP), engine tampering levels, inspection and maintenance programs, and vehicle emission standards. If emissions are being calculated for a specific day or month, MOBILE also requires input information for factors such as maximum and minimum temperature and sunrise and sunset times.
11. MOBILE6 produces emission factors and VMT weighting factors, typically for each county, urban/rural area, and roadway functional type. VMT is multiplied by the appropriate emission factors to determine emissions.

Note that the standard on-road vehicle emission inventory process using the MOBILE model, as outlined above, does not calculate emissions from long-term truck idling. A small amount of idling emissions are implicitly calculated by virtue of the standard urban driving cycle used in emission factor development (which include some stop-and-go driving patterns). The next generation EPA emissions model (MOVES) is expected to more fully capture long-term idling. In California, the EMFAC model has recently been modified to calculate some long-term idling emissions associated with each truck trip.

Summary of Results

Heavy-duty trucks account for 6.0 to 12.7 percent of total on-road VMT in the six study regions. The truck fraction is highest in the Detroit region (12.7 percent), probably due in part to the high U.S.-Canada truck volumes in the region. The Los Angeles region has the largest total heavy-duty truck VMT, but the smallest truck VMT fraction (6.0 percent). This is likely because the very large regional population and passenger vehicle activity means that trucks are responsible for a relatively small share of the on-road total.

Table 3-3 presents a comparison of heavy-duty truck emissions in the six study regions. Numbers in bold reflect the highest truck percentage among the six regions. The regions show some significant differences in terms of the relative contribution of trucks to total on-road pollutant emissions. The contribution of freight trucks to total on-road NOx emissions ranges from a high of 63 percent (Detroit) to a low of 49 percent (Los Angeles). The contribution of freight trucks to total on-road PM-10 emissions ranges from a high of 63 percent (Chicago) to a low of 31 percent (Los Angeles). (Two regions did not provide total on-road PM-10 emissions.)

Table 3-3: Comparison of Heavy-Duty Truck Emissions in the Six Study Regions, 2002

Source: Based primarily on data provided by state air quality agencies and MPOs; see report text for details.

A number of factors contribute to these differences. Some industry sectors are more transportation intensive than others, so differences in regional economic structure create different levels of trucking activity. Differences between Los Angeles and the other regions are caused in part by differences in the MOBILE6 and EMFAC emission factors. Some of the emissions differences may also be caused by differences in the composition of the truck fleet. For example, in the Los Angeles region, gasoline trucks account for the largest share of total truck VMT (32 percent), which means that truck VOC and CO emissions are relatively larger in Los Angeles. Because long-haul trucks tend to be larger combination vehicles, regions with more pass-through truck traffic will tend to have a larger share of Class 8b truck VMT (diesel-powered) and therefore, will have higher NOx and PM emissions.

3.3 Freight Railroad Emissions

The standard approach for calculating railroad emissions is generally the most simplistic, and potentially the least accurate, of the four major freight modes. Unlike trucks, marine vessels, and aircraft, which use publicly-owned facilities, there is typically little or no published information on private railroad activity available for a specific region. Thus, state and regional air quality agencies must rely on obtaining railroad activity data directly from the railroad companies. Even when this data is provided, it is often not reported with a high level of detail, due in part to the railroad company procedures for maintaining this data.

To determine freight rail emissions in the six study regions, we relied on data provided by state air quality agencies. In some cases, we modified or supplemented state emissions estimates; in other cases, we report the state figures as provided.³⁶

Summary of Methodology

All six study regions follow a similar methodology to estimate railroad emissions. This involves estimating county-level fuel use for line-haul locomotives and, separately, for switch yard locomotives. Fuel use estimates are then used to calculate emissions. The steps in this approach are outlined below. The details of the methodology (and its accuracy) depend heavily on the nature of the locomotive activity data provided to the states by the railroads.

1. Each freight railroad that operates in a state is asked to report their gross ton-miles (GTM) by county, as well as their total fuel consumption in the state. If a railroad is able to provide this information, the statewide line-haul fuel use is apportioned to counties in direct proportion to the GTM. Sometimes the railroads perform this fuel use allocation, using their own estimate of fuel use per GTM.
2. Some railroads are not able to report GTM. For these railroads, mileage of active track is used as a proxy. If the railroad is able to report statewide line-haul fuel use, fuel use is apportioned to counties in direct proportion to the railroad's track mileage by county. If the railroad cannot report statewide fuel use, national-level fuel use (as reported by the Association of American Railroads) is apportioned to state and county based on track mileage.
3. Each freight railroad that operates in a state is asked to report the number of switch yard locomotives they operate, by county or by individual yard. Some railroads are able to provide hours of switch locomotive use by county or yard. Railroads are also asked to report the average annual fuel consumption rate (in gallons per locomotive per year) of their switch yard locomotives. If railroads cannot provide this rate, a rate is assumed based on EPA guidance or information from other railroads. Switch yard locomotive fuel use is then calculated by applying a fuel consumption rate to the number of switch yard locomotives.
4. Class II and III railroads (shortline and switching railroads) are often unable to provide the information described above. In some regions (such as Chicago), the number of Class II/III railroads in operation is considered too large to make surveys of individual companies practical. In these cases, fuel consumption can be estimated by obtaining the number of employees of the railroad by county (using a commercial employment database such as Dun & Bradstreet) and a ratio of fuel consumption per employee.
5. The fuel use estimates for each railroad are summed. The result is an estimate of total railroad fuel use by county.
6. Emission factors (in grams per gallon) are applied to the fuel use figures to estimate annual emissions.

There are a number of potential shortfalls to this methodology. Most notably, length of active track is almost certainly not an accurate proxy for fuel use. In most regions, some rail lines are used much more heavily than others. Thus, using track length to apportion fuel consumption to the county level probably results in significant inaccuracies. GTM is a much better proxy for fuel use, but there have been questions about the accuracy of county-level GTM data reported by railroads.

There have also been questions about the accuracy of the fuel consumption data reported by railroads.³⁷ For example, the fuel use reported by railroads for Texas' 2001 inventory (220 million gallons) is less than half the locomotive fuel sales for the state as reported by the U.S. Department of Energy (504 million gallons) for that year. Some of this discrepancy can likely be explained by the fact that railroads often purchase fuel in one state and then consume that fuel in another. Unfortunately, there are no mechanisms to verify the fuel consumption data reported by railroads.

Some states use locomotive emission factors from EPA's 1992 emission inventory guidance.³⁸ These emission factors are likely outdated. In cases where we were able to obtain fuel use data (Baltimore, Dallas, Houston), we calculated emissions using 2002 emission factors provided in EPA's 1998 Regulatory Support Document, which was developed to support the adoption of locomotive emission standards.³⁹

Summary of Results

Table 3-4 shows the freight rail emissions totals in the six study areas. Chicago has far more freight rail emissions than any other region - approximately twice the emissions in the Los Angeles region (with the exception of CO) and more than four times (in some cases, more than 10 times) that in any of the other regions.

Table 3-4: Freight Rail Emissions in the Six Study Areas, 2002

Source: Based primarily on data provided by state air quality agencies; see report text for details.

The regions show significant differences in the contribution of switch yard locomotive activity to the freight rail emissions total, as shown in Table 3-5. In Baltimore, for example, switchers are estimated to be responsible for more than half of the freight rail emissions. In contrast, switcher locomotives in the Los Angeles region contribute only 10 percent of NOx and 8 percent of PM-10 from freight rail, according to the region's emission inventory. Some of these differences are likely a product of variations in the inventory development methods.

Table 3-5: Contribution of Yard Operations to Freight Rail Emissions Total

Source: Based primarily on data provided by state air quality agencies; see report text for details.

3.4 Marine Freight Emissions

Marine freight sector emissions are caused by the engines used to power vessels and associated equipment and by engines in the land-based equipment that are used for handling marine cargo at ports. Marine freight includes shipping to and from U.S. coastal ports, in the Great Lakes, and in navigable inland waterways. Freight shipping vessels range from non-self-propelled barges and scows to self-propelled container ships, bulk carriers, tankers, and tugboats.

Land-based port emissions originate from three general sources: on-dock equipment, trucks, and locomotives. Emissions from on-road trucks and locomotives at ports are captured in the estimates of the emissions from the trucking and railroad sectors, respectively, so we have not included them in the marine freight sector. On-dock equipment includes the equipment used to load and unload freight from ships, service the ships, and move freight within the port area. We refer to this equipment as cargo handling equipment and have developed estimates of these emissions for each port. Examples of cargo handling equipment include yard tractors (or yard hostlers), top and side loaders, forklifts, and cranes.

As with all transportation sources, estimating emissions from marine freight generally involves multiplying an activity parameter by an emission factor. Activity for marine vessels is typically described in terms of tonnage, calls, or trips at a port in a given period, usually one year. Tonnage is the mass of goods loaded or unloaded at a port. A call is a single entrance and exit from the port boundary. A trip is a single movement of a vessel and can include the movement into a port, the movement out of a port, or a vessel shift within a port. Thus, vessel trips at a given port are always two or more times the number of calls. The U.S. Army Corps of Engineers' Waterborne Commerce of the United States series reports annual tonnage and trips for every port in the U.S. More detailed data can be obtained from individual ports.

Summary of Methodology

Development of an accurate marine vessel emission inventory also requires information on the time vessels spend in different operating modes in the port region, typically the following:

• cruise - operating at sea speed
• reduced speed zone (RSZ) - speed roughly 30 percent of cruise, typically occurring between the breakwater and the maneuvering zone
• maneuvering - dead slow or reverse, typically about 3-4 knots occurring within about 2 miles before the vessel reaches its dock or anchors
• hotelling - time spent at dock or anchorage

Several parameters for each ship type must be specified in order to properly characterize emission rates. These include the total power of the main engines, the load factor (the fraction of full power used in each operating mode), and the power of the auxiliary engines.

The Port of Los Angeles and Houston have recently developed marine vessel emission inventories, and we report these emissions estimates, scaling to 2002 as necessary. For the other ports in the study regions, we estimated vessel emissions using a combination of EPA guidance, methodologies and data from other studies, and published current port activity data. Our inventories for the ports of Baltimore, Chicago, and Detroit are developed primarily based on the methodologies laid out in reports for EPA by ARCADIS^{40, 41} and Environ⁴² and use parameters developed for the most recent emissions study for the Port of Los Angeles.⁴³ For the Port of Long Beach, we estimated vessel emissions based on the Port of Los Angeles inventory and the ratios of cargo tonnage between the two ports.

No EPA guidance or other standardized methodology exists for developing estimates of port cargo handling equipment (CHE) emissions. For the ports of Houston, Los Angeles, and Long Beach, we used the recent CHE inventories developed by the ports.^{44, 45, 46} For the other study ports, we developed a methodology that relies on the Los Angeles and Long Beach CHE emission inventories and scales emissions using appropriate cargo tonnage. Table 3-6 summarizes the methods used for estimating vessel and CHE emissions at the study ports.⁴⁷

Table 3-6: Summary of Emissions Estimation Process

Summary of Results

Table 3-7 shows total marine freight vessel and port CHE emissions in the study area ports. The Los Angeles region has by far the greatest marine freight emissions - more than 22,600 tons of NOx and more than 1,500 tons of PM-10 annually. The Houston metropolitan area has more than 14,000 tons of NOx and more than 900 tons of PM-10 annually from marine freight. Marine freight emissions in the other three regions are smaller - roughly 3,300 tons of NOx and 190 tons of PM-10 in Baltimore, 2,200 tons of NOx and 175 tons of PM-10 in Chicago, and 500 tons of NOx and 30 tons of PM-10 in Detroit.

The Port of Houston has the greatest marine vessel emissions of any single port, followed closely by the Ports of Long Beach and Los Angeles. Port CHE emissions are greatest at the Port of Long Beach, followed by the Port of Los Angeles. CHE emissions make up approximately 20 percent of the marine freight total at these ports. At the Port of Houston, CHE emissions are only about 10 percent of the marine freight total. This difference reflects differences in the freight handled at the ports - Houston handles a large proportion of liquid bulk freight (mostly petroleum), which requires relatively little in terms of land-side CHE, while Los Angeles and Long Beach handle large volumes of containers, which require extensive land-side activity by CHE.

Table 3-7: Total Marine Freight Vessel and Port CHE Emissions by Port

Source: Based on Port emission inventories and calculations by ICF Consulting; see report text for details.

Table 3-8 shows a comparison of NOx emissions by vessel type. This comparison illustrates the vast differences in vessel and cargo type between study area ports. Containerships are the largest single emissions source at the Ports of Los Angeles and Long Beach. At the Port of Houston, 40 percent of NOx emissions come from tankers, while tankers contribute relatively little at the other ports. Emissions at the Ports of Chicago and Detroit are dominated by bulk carriers. Tugs and harbor craft account for significant portions of NOx emissions at most of the ports and approximately one-third of all vessel emissions at Los Angeles and Long Beach.

Table 3-8: Comparison of Marine Freight NOx Emissions by Vessel Type

Source: Based on Port emission inventories and calculations by ICF Consulting; see report text for details.

Table 3-9 shows a comparison of NOx emissions from hotelling by ocean-going vessels (i.e., excluding tugs and other harborcraft) at the study area ports. The contribution of hotelling to total ocean-going vessel (OGV) emissions varies significantly. It is highest at the Texas ports and at the Port of Baltimore. Hotelling accounts for roughly 30 percent of OGV emissions at Los Angeles and Long Beach. Hotelling contributes very little to OGV emissions at the Ports of Chicago and Detroit.

Table 3-9: Comparison of Marine Freight OGV Hotelling Emissions

Source: Based on Port emission inventories and calculations by ICF Consulting; see report text for details.

Table 3-10 shows a comparison of the NOx emissions from port CHE for the three ports that were able to provide CHE emissions by equipment type. Yard tractors make up the largest component of port CHE emissions in all cases. This comparison shows that, while yard tractor emissions are similar at the Ports of Los Angeles and Long Beach, emissions from handlers/loaders and from cranes are significantly higher at Long Beach. Emissions from yard tractors and handlers/loaders are relatively smaller at the Port of Houston than at the San Pedro Bay ports, reflecting the relatively small share of containerized cargo at Houston.

Table 3-10: Comparison of Port CHE NOx Emissions by Port

Source: Based on Port emission inventories and calculations by ICF Consulting; see report text for details.

3.5 Air Freight Emissions

In this section, we present estimates of the emissions that are attributable to air transport of freight. Air transport is by far the smallest of the four freight modes on a tonnage or ton-mile basis. Nationally, air freight accounts for 0.4 percent of domestic freight ton-miles. Air freight is the most rapidly growing freight mode, however, with ton-miles nearly doubling since 1990.

Emissions from air freight are generated by aircraft and by airport ground support equipment (GSE). Aircraft include those devoted exclusively to cargo and passenger aircraft that carry freight together with passenger baggage in the cargo space (belly cargo). Airport GSE include aircraft and baggage tow tractors, ground power units, portable aircraft air conditioning units, and air start units, as well as medium and light duty trucks for such operations as refueling and de-icing. Over the past five years, many of the nation's airports have been working to electrify GSE, and as a result, GSE emissions are dropping significantly. For this study, we have estimated GSE emissions attributable to freight based on the same fractions used in determining the aircraft emissions associated with freight, as discussed below.

Summary of Methodology

Aircraft emit pollutants during flight that, due to atmospheric mixing, affect ground level pollutant concentrations. This mixing zone extends to 3,000 feet on average, and all pollution emissions in this zone are included in an airport emission inventory. The aircraft operations of interest within the mixing zone are defined as those in the landing and takeoff (LTO) cycle. Exhaust emissions are calculated for one complete LTO cycle of each aircraft type using emissions factors for the aircraft's specific engines at each power setting or mode of operation, as well as the time spent in each mode. The activity of aircraft for the inventory period can then be multiplied by emission factors to calculate the total emissions.

All regions use the FAA-sponsored Emissions and Dispersion Modeling System (EDMS) to develop emission inventories for airports. For each of the major freight airports in the six study areas, we obtained from the state or regional air quality agency an annual emission estimates from the most recent application of the EDMS model. We considered the airports of interest to be those handling at least five percent of a region's air cargo, which includes the major passenger airport in each region plus the Ontario airport in the Los Angeles region and the Alliance airport in the Dallas-Fort Worth region.

The EDMS model does not distinguish between freight movement from non-freight (passenger) movement. Therefore, we developed an approach to allocate each airport's total commercial aircraft emissions to the freight and non-freight sectors, as described previously in Section 2.1 and summarized below.⁴⁸

We obtained commercial aircraft departure records from the BTS Air Carrier Statistics database. We assumed emissions from air cargo aircraft are attributable entirely to freight. Alliance Airport in Fort Worth (AFW) is unusual in that it handles air cargo almost exclusively. Among the other study airports, air cargo aircraft account for 0.4 to 7.0 percent of departures. For passenger aircraft departures, we estimated the weight of the aircraft's freight (belly cargo) and the weight of the aircraft's passengers and baggage and used these percentages to allocate emissions.⁴⁹ Through this process, we estimate that freight is responsible for 1.4 to 6.7 percent of passenger aircraft emissions at the study airports. In total, we estimate that the air freight share of aircraft emissions ranges from a low of 2.3 percent at Detroit to a high of 10.9 percent at LAX, among the study airports with passenger service (the exception being AFW, which is nearly 100 percent freight).

Summary of Results

Table 3-11 summarizes the emissions attributable to air freight movement for five criteria pollutants: VOC, NOx, CO, SOx, and PM-10.⁵⁰ LAX has by far the largest freight emissions among the eight study airports. Aircraft emissions of VOC, NOx, and CO in the Los Angeles region are approximately 70 percent more than the next largest region (Chicago). Emissions of SOx (and PM-10, which we calculated from the SOx emissions) are proportionally lower for the Los Angeles airports, most likely due to the use of jet aircraft fuel with lower sulfur content, as required in California.

Table 3-11: Aircraft Emissions Associated with Freight Movement, 2002

Source: Compiled by ICF Consulting based primarily on data provided by state and regional air quality agencies; see report text for details.

Table 3-12 shows GSE emissions attributable to air freight movement for five criteria pollutants: VOC, NOx, CO, SOx, and PM-10. We obtained GSE emissions from each state or regional air quality agency and allocated a portion to freight using the same methodology and ratios as discussed above for aircraft. With the exception of Baltimore, all of the air quality agencies estimated the emissions for ground support equipment outside of the EDMS model, which likely reflects past problems with the emission factors used in the EDMS. The latest version of EDMS uses the same emission factors as the current EPA NONROAD model. GSE emissions for DTW and AFW were not available from air quality agencies. For DTW, we estimated GSE emissions by multiplying ORD GSE emissions by the ratio of air freight activity at DTW to ORD. For AFW, we estimated GSE emissions by multiplying the DFW GSE emissions by the ratio of air freight activity at AFW to DFW.

The results show that LAX has again by far the largest air freight-related emissions of NOx and PM from GSE among the eight study airports. In all airports, NOx and PM-10 emissions from GSE are generally much less than the freight aircraft emissions of those pollutants, typically less than 20 percent of the aircraft emissions. The high CO emissions reflect the use of gasoline fuel in much of the ground support equipment.

Table 3-12: Ground Support Equipment Emissions Associated with Freight Movement, 2002

Source: Compiled and calculated by ICF Consulting, based primarily on data provided by state and regional air quality agencies; see report text for details.

3.6 Summary and Comparison

Table 3-13 shows a comparison of freight transportation NOx emissions by mode. Emissions are greatest in magnitude in Los Angeles, followed by Chicago and Detroit. NOx emissions from freight in the Los Angeles region are nearly five times those in Baltimore and nearly three times those in Dallas-Fort Worth.

Table 3-13 clearly shows the dominant role of trucking in urban freight movement and emissions. Heavy-duty trucks are responsible for more than three-quarters of freight emissions in all six regions. In Detroit and Dallas-Fort Worth, trucking accounts for virtually all freight emissions - 97 percent of the freight total in Detroit and 93 percent in Dallas-Fort Worth.

In other modes, the regions show considerable diversity in terms of freight emissions. Freight rail NOx emissions in Chicago are nearly twice that in any other region and make up almost 20 percent of Chicago's total freight emissions. In the other five regions, freight rail accounts for less than 10 percent of the total.

Marine freight NOx emissions are greatest in the Los Angeles region, where they account for 14 percent of the freight total, and in Houston, where they account for 17 percent of the total. Air freight emissions are dwarfed by the other modes in all six regions. Air freight NOx emissions are greatest in the Los Angeles region, making up 0.5 percent of the region's freight total.

Table 3-13: Regional NOx Emissions from Freight by Mode, 2002

Source: Compiled and calculated by ICF Consulting, based primarily on data provided by state and regional air quality agencies, MPOs, and ports; see report text for details.

Table 3-14 shows the comparison of freight transportation PM-10 emissions across modes. Trucking is still the largest contributor, though less dominant than with NOx emissions. In particular, marine freight accounts for a major portion of freight PM-10 emissions in regions with large seaports - 40 percent of the total in Houston, 37 percent in Los Angeles, and 19 percent in Baltimore. This in part reflects the high PM emission rates of large marine vessels that burn residual fuel and have little or no emission controls.

Table 3-14: Regional PM-10 Emissions from Freight by Mode, 2002

Source: Compiled and calculated by ICF Consulting, based primarily on data provided by state and regional air quality agencies, MPOs, and ports; see report text for details.

Table 3-15 compares annual freight NOx emissions with total mobile source and total emissions (mobile, area, and point sources). Freight accounts for 40 to 52 percent of all mobile source NOx emissions, and 29 to 39 percent of all NOx emissions in the study regions (total emissions data was not available for Baltimore). This is significantly higher than the national freight share of NOx emissions (26.8 percent) presented in Table 2-9. Freight NOx emissions are highest in absolute terms and in percentage terms in the Los Angeles region, which likely reflects the large contribution from the region's ports.

Table 3-15: Regional Freight NOx Emissions Compared to Total Mobile Source and Total Emissions, 2002

Source: Freight emissions from sources as described in report text. Total mobile source emissions and total emissions obtained from state air quality agencies; data in most cases reflects preliminary submittal of 2002 emission inventory data under EPA's Consolidated Emissions Reporting Rule.

Table 3-16 compares annual freight PM-10 emissions with total mobile source and total emissions (mobile, area, and point sources). Freight accounts for 22 to 47 percent of PM-10 emissions from mobile sources in the study regions. Compared to total emissions, freight accounts for 1.0 to 5.8 percent of PM-10 emissions in the study regions. Again, this is higher than the national freight share (0.8 percent) presented in Table 2-9. Freight accounts for the largest share of total PM-10 emissions in the Chicago region, which likely reflects the intensive railroad activity there. Note, however, that the vast majority of PM-10 emissions come from agricultural fields, wildfires, and fugitive dust. The total PM-10 emissions in the six regions, and the portions attributable to freight, therefore depend heavily on the amount of undeveloped land within the nonattainment boundaries. Note also that the PM emissions from freight transportation are a greater concern than the coarse particulates from sources like fugitive road dust. Current emission inventories do not provide an accurate estimate of fine particulates, so it is difficult to assess the freight sector contribution to these emissions.

Table 3-16: Regional Freight PM-10 Emissions Compared to Total Mobile Source and Total Emissions, 2002

Source: Freight emissions from sources as described in report text. Total mobile source emissions and total emissions obtained from state air quality agencies; data in most cases reflects preliminary submittal of 2002 emission inventory data under EPA's Consolidated Emissions Reporting Rule.